Predicting the conversion from clinically isolated syndrome to multiple sclerosis: An explainable machine learning approach.

INTRODUCTION: Predicting the conversion of clinically isolated syndrome (CIS) to clinically definite multiple sclerosis (CDMS) is critical to personalizing treatment planning and benefits for patients. The aim of this study is to develop an explainable machine learning (ML) model for predicting this conversion based on demographic, clinical, and imaging data. METHOD: The ML model, Extreme Gradient Boosting (XGBoost), was employed on the public dataset of 273 Mexican mestizo CIS patients with 10-year follow-up. The data was divided into a training set for cross-validation and feature selection, and a holdout test set for final testing. Feature importance was determined using the SHapley Additive Explanations library (SHAP). Then, two experiments were conducted to optimize the model's performance by selectively adding variables and selecting the most contributive variables for the final model. RESULTS: Nine variables including age, gender, schooling, motor symptoms, infratentorial and periventricular lesion at imaging, oligoclonal band in cerebrospinal fluid, lesion and symptoms types were significant. The model achieved an accuracy of 83.6 %, AUC of 91.8 %, sensitivity of 83.9 %, and specificity of 83.4 % in cross-validation. In the final testing, the model achieved an accuracy of 78.3 %, AUC of 85.8 %, sensitivity of 75 %, and specificity of 81.1 %. Finally, a web-based demo of the model was created for testing purposes. CONCLUSION: The model, focusing on feature selection and interpretability, effectively stratifies risk for treatment decisions and disability prevention in MS patients. It provides a numerical risk estimate for CDMS conversion, enhancing transparency in clinical decision-making and aiding in patient care.

Topical Dorzolamide as Adjunctive Treatment With Intravitreal Bevacizumab in Bilateral Diabetic Macular Edema.

BACKGROUND:  Intravitreal injection of anti-vascular endothelial growth factor (VEGF) agents is accepted as the gold standard treatment for center-involving diabetic macular edema (CI-DME). Adjunctive administration of topical dorzolamide may enhance the therapeutic effects of anti-VEGF agents. In this study, we compared the efficacy of topical dorzolamide plus intravitreal injection of bevacizumab (IVB) versus IVB alone in patients with bilateral DME. METHODS: This prospective, randomized contralateral eye study was carried out in a tertiary referral ophthalmology center, Al-Zahra Eye Hospital, Zahedan, Iran, between April 2021 and April 2022. This study included 50 eyes of 25 patients with bilateral DME. All eyes received three consecutive monthly injections of IVB. For each patient, one eye was randomized to instill dorzolamide eye drops three times a day as an intervention, and the other received artificial tear drops as a placebo. Best-corrected visual acuity (BCVA), central macular thickness (CMT), and intraocular pressure (IOP) were evaluated before starting treatment and then monthly for the first three months. RESULTS: Among 25 included patients, the average age was 56.64 ± 7.97 years, and 48% were female. BCVA did not improve significantly in any groups (P > 0.05). No significant difference was observed in terms of BCVA between the intervention and control groups (P > 0.05). The present study showed a decrease in CMT in both study groups (P < 0.05). At month 3, the decrease in mean CMT from baseline was significantly higher in eyes receiving topical dorzolamide compared to the control group (-88.92 ± 82.90 vs. -37.64 ± 86.16 µM, respectively; P = 0.037). IOP decreased significantly only in eyes receiving dorzolamide (P < 0.001). CONCLUSIONS: The results of the present study indicate that adjunctive administration of topical dorzolamide has a beneficial effect on CMT reduction from baseline, but it did not have an additive effect on BCVA improvement compared to IVB monotherapy.